If this is your first visit, be sure to
check out the FAQ by clicking the
link above. You may have to register
before you can post: click the register link above to proceed. To start viewing messages,
select the forum that you want to visit from the selection below.
No announcement yet.
What Linux Distribution Should Be Benchmarked The Most?
Enterprise means old, stable. The penalty for old and stable is usually slower performance (less optimizations). Maybe this doesn't hold true across the whole software stack, but for video drivers it usually does.
Probably yes, but it seems by default they use debug build on bleeding edge packages, and this might lead to misleading results (everything would look slower than it would really be when it comes to end users).
I know that ubuntu is the most used (30 millions of users) and i don't know which other distro recommend to the noobs. But Ubuntu don't have to be more considered a genuine linux distro. Canonical has broke the rules there must be consequences. Cannot recommend Ubuntu derived (Mint, Xubuntu, etc) because are still ubuntu. Maybe should search for another debian derived, like Linux Mint Debian Edition or something.
I never claimed that Ubuntu was "needed by the rest of the "GNU/Linux platform" - obviously the kernel and GNU tools would continue to exist without Ubuntu, as would every distribution that is not based on Ubuntu. What I was pointing out is that there are many derivatives of Ubuntu, in response to the comment that disagreed with "Be careful what u wish for, if ubuntu would to fall it wont fall alone...". Yeah, if Ubuntu falls then there would be a bunch of Ubuntu based distributions that would fall too. Their users would suddenly find that they can't upgrade, and aren't getting any more bug or security fixes. The world would survive and people would move on, either switching to other distributions or Windows, but in the end the failure of one of the most popular and high profile Linux distributions would be incredibly bad PR for the Linux community. We need more successful Linux-based companies, not less. And if millions of Linux users are suddenly left without upgrades, security and bug fixes, then you can bet that it will be remembered for a long time, and Apple/Microsoft will be gloating that "You can't rely on Linux, your distribution can just disappear without any warning, leaving you to clean up the mess".
Oh. Well, your first post was really ambiguous in that regard, but good to know what you actually meant by that. But Ubuntu is not going anywhere any time soon, in any case.
Reading all the comments, about who's distro is the best. Perhapse the best idea for benchmark is to benchmark the foundation distros instead of the child ones.
Debian 7 (no GUI) instead of Ubuntu.
Redhat instead of say OpenSuse.
The child distros add and tweak their linuxes, or add or subtract things that may skew results. You need a baseline before you can compare.
So I'd advise to benchmark Debian 7, right from debian.org. Then compare it to Ubuntu. The same with Redhat as well.
I look forward to the results then.
No, that still doesn't make much sense. Both Debian and CentOS are very stable, which means that they don't reflect the current state. Also, openSUSE is a foundation distribution – it started off as a fork of Slackware, but hasn't relied on it for many years now. Fedora is likewise a foundation distribution. It started off as a fork of Red Hat Linux, but it's been the driving force behind RHEL ever since, not the other way around. So while I would agree that the more independent distributions should be tested, they shouldn't be oldly stable. Debian testing would be fine in that regard, though.
Also Michael, you should get special sponsoring from the likes of Red Hat and Canonical, if they want to always appear in the benchmarks no matter their popularity...
Sysadmins might want to know what they trade in production performance when switching from red hat to ubuntu server, etc.
And workstation benchmarks need timed tests, from boot to login, login to desktop, desktop to launching libreoffice, stuff like that (with macro like movement of cursor and pushing things to launch apps, etc). This will show the "snappiness" of the different desktops and can help to figure out if a real time kernel; or changing the scheduler helps or harms, etc.
Oh and power consumption and temp readings for the portable and datacenter people.
Well i suppose you could stick to the 3 most popular community distros + sponsored or such.
There is no need for loyalty to a for profit company that is not contributing to your site.